February 2005 Archives

“End users want to achieve their goals with minimum of cognitive load and a maximum of enjoyment.” ~ Marchionini. Why? Because search users are nitwits. Mike asks us to consider the following. What if someone goes into a travel store and when asked what he is looking for, he answers “travel”. He goes on to describe it takes to get ranked in the top ten. Social sciences and bibliometry is also mentioned on the screen and have existence for a long time, even before search engines. They are being applied today in the algorithms that are created for search engines. The web is a social network he continues. Social networks have been extensively researched long before the web. He describes citation analysis and the how this is applied to in search engines. There is a difference between a citation and a reference.

Hyperlink analysis algorithms make either one or both of these simple assumptions. Assumption 1 – A hyperlink from page A to page B. Co citations, if a page C cites pages A and B, then A and B are said to be co-cited by C. Pages A and B being co-cited by many other pages is evidence. There are two main algorithms based on links. PageRank (Google): Each page on the web has a measure of prestige that is independent of any information need or query i.e. keyword independent. Roughly speaking, the prestige of a page is proportional to the prestige of the sum of the prestige scores of pages. HITS or Hyperlink-Induced Topic Search. Problem is that neither of these algorithms work.

The problem with HITS. Topic drift, nepotistic linking, and runtime analysis. Mike says there are three steps to success. They cracked the problem relating to time of a search from 11 seconds to instant. He describes Teoma and subject specific popularity.
Adventures in search algorithms: What happened next? Both Krishna Bharat and Monica Hensinger join Google. Mike believes that Florida that moved from keyword independent to keyword dependent.
Ending joke:
There is a guy trapped in the desert and is looking for life. He finds a man face down in the sand, with a bag on his back. He thinks what was in the bag that would have saved him. Answer: Parachute

Next up was Rahul Lahiri he presents some of the properties that Ask Jeeves controls. Today they are ranked #7 on the web and have done exceedingly well since this time last year. What is their mission: relevance. He goes into general link analysis methods. The challenge is to discovering what the links are about. A link from page A to page B (or C) is a vote or recommendation by the author or page A for the page B (or C). The problem is that if you have a link with the anchor text budget, you don’t know what the budget means. Was it a budget for Budget rent-a-car or budget for someone’s companies?? That’s a problem obviously. He continues that organizing into local subject communities of sites. This is how Teoma views that web. Some of the challenges that they face is that solving the problem in real-time. 200 ms (milliseconds) to do this computation for each query, millions of times per day. You also have to identify the communities. The link structure of the web is noisy. Hubs link to topic specific pages. An example of topic focused vs. broad topic areas. Topic focused is a search for “buffalo” and broad topic areas is a search for “bay area airports”. Some of the benefits are that smaller enthusiast sites get a chance to come up to the top of the search listings (example search: fantasy football).
The power of communities is a better vision, expert validation, contextualization, and better user experience.

Next Dr. E. Garcia, a pioneer that has allowed us to better understand the search engines as marketers was next to present. His plane has been delayed till tomorrow because of weather (its snowing heavily here), BUT there is a voice over for his presentation. Tapes starts. He is going to discuss grasping co-occurrence. Co-occurrence suggests association of relatedness. Side note: People are leaving because the audio isn’t too great. But not too many as there is a good amount of interest for this. Back to co-occurrence. Co-occurrence can be: Global, Local, or Fractal. This presentation is highly technical, and while I understand his work, it’s hard to follow. I am trying to get what I can, as its requiring very detailed listening and comprehension at this point. I apologize for any errors in this document.

Example of the case of “Hawaii” which is semantically connected to aloha, Hawaiian, Maui. C-indices can be used to estimate the relative presence of targeted keywords across search engines. He gives another example of “comida + mexicana” that are semantically connected. Example: C-indices can be used to monitor keyword trends, word patterns and topics in time. He goes on to talk about competitive words. Based on his research the example suggest that many competitive queries in Google tend to exhibit C12 indices. His research indicates that overused queries tend to exhibit unusually high C-indices while unrelated terms in a query tend to exhibit very small c-indexes. He gives the example of “guacamole optimization” with a low c-index of 0.12. On to term sequencing: EF-ratios. He talks about various types of queries such as a findall and exact and how order and frequency matter. He goes on to give the example that EF-ratios can be used to estimate the relative frequency of natural sequences and phrases in a source. So what about candidate sequences? These EF ratios can be used to examine how easy or difficult would be to rank for a given sequence in a given search. Keyword competitiveness is specific to each search engine. Some search engines return documents whose sequence can be found. When queried in EXACT mode, some searches return docs in which the queried term can be found. What is it separated by, delimiter (hyphen, underscore), space, or stopwords (in, of, with). So to recap, co-occurrence theory can be used to understand semantic associations between: terms, products, services.

Q: Interested in how we will be searching in 5-10 years time? Personalization?
A: Where is search going? Mike did an interview from the founder of Teoma. It was interesting he says. The most interesting is that he said they need to get up 10 steps up the ladder, currently we are 3-4. The one thing that will change this, will be personalization. It’s misunderstood, personalization. It’s not giving you a search just for you. Its about returning results for your peer group. They can start to tailor the search specifically to you. There is data now using genetic algorithms and others set that are using these to create search engines. Mike concludes the more information we give the search engines, the better our experience will be.

Orion couldn't make it due to a mud slide, he will be here tomorrow. They will try to present his presentation with voice overs.

Mike Grehan was up first, he deleted his presentation last night. So he had to restart from scratch, but everyone sympathized. He shows the SEW Forums, and explains that people are very interested in "this stuff." He highlights the keywords co-occurance thread that had 46,401 views, so there is a lot of interest. He said Orion deserves a ton of credit. What are the ages of my three sons? He starts a story that all of his three sons are having a birthday today. He then gives clues to figure out his sons ages, the product of the ages of my sons is 36 and the sum of their ages is equal to the number of windows in the building and the last clue is one son has blue eyes. Mike then gives down a break down on how to figure with equations. the answer is 9, 2, and 2. The last clue, about the blue eyes, said there was an oldest son so the 6, 6, and 1 wouldn't be the right answer. He explains that engines want the most relevant results, which is hard "because end users are search nitwits!" He explained that someone who walks into a travel store and tells the clerk "travel" he will kick you out but search engines respond. The "abundance" problem, too many results, which are the best results, which are the most relevant? Social networks have been extensively researched long before the Web. He briefly explains "Citation analysis", so we have a Web graphic, directed edges and undirected edges (co-citation). If you have questions about this, let me know. Then he discusses PageRank and HITS. PageRank he sums up, PageRank is keyword independent. HITS (Teoma) which is keyword dependent. Great way of explaining the difference. He says there is only one problem with these two solutions, "Neither of them work." He said the problem with PageRank, well they don't use it, so he skipped it. He then went on to HITS and said topic drift, nepotistic linking and runtime analysis are the three issues. The first two were corrected, but runtime analysis is still an issue. He said how AG from Ask Jeeves (Teoma) cracked it. He then put up a graph on the hubs and authorities. So what happened next? B&H algorithm died with AV, then those two went to Google and Hilltop came out. Then in Feb. 03, Google patented Local Link (Bharat). Then he went into Florida (nice little graph), he said it had a lot to do with Google moving from keyword independent to dependent. He throws up some links to advanced papers on this about the future. He finishes off his presentation with an other story. A guy is walking in a desert, he finds a dead guy on the sand with a bag on his back. What was in his bag, a parachute.

Next up was Ask Jeeves named Rahul Lahiri, he helped me out once with a relevancy issue a month ago. He said there is some overlap with Mike's presentation. He goes over the Ask properties and growth numbers. Ask's mission is relevance, index completeness, freshness, and structured data (smart answers). Algorithmic drives are content/text analysis, and link analysis. He focuses on the link side; and shows a graph of page a linking to page b and page c (mike showed something similar). Ask looks at what the "links are about". He goes into the hubs and authority thing. The key challenges are solving the problem in real-time and identifying the communities. He then gives examples of queries such as "buffalo" vs. "bay area airports". They need to weed out the noises from the good stuff. He explains that small enthusiast sites get a chance to rise to the top, which is great. They then can do a better job of identifying different communities, refine search.

Now they give Orion's, Dr. E. Garcia's presentation a try. It sounds like Nacho. Cool, its working. Nacho introduces it. Co-occurrence suggests association or relatedness. I'll summarize it later, very technical.

UPDATE:
First excuse me if I make major mistakes in my interpretation of the presentation. I hope Dr. G. (Orion) reviews this and makes any necessary corrections.

Orion's first slide went over some of the basics of co-occurences. Orion explains that co-occurences shows a type of "relatedness" between words. So if you have two terms that are often discussed or found on the same document, they tend to be more related. He then gives an example of the term "aloha". What does aloha make us think of? Hawaii is the correct answer. Orion then explains that this is important when conducting "keyword-brand associations." In Orion's second example he shows an equation he discussed in the forums; c12-index = (n12/(n1+n2-n12))x1000, he overlays an example of a k1 and k2 showing the n12 overlap in the middle as well as explains how an example of 3 keywords makes for a much more complex query in AND mode (n123). He then brings back the old example of "aloha hawaii" to explain "term associations". When you compute the values in Google of "aloha hawaii" versus "aloha indiana" or "aloha montana" you will notice the the C index is much higher with "aloha hawaii" (28.11) versus "aloha indiana" (3.23). This shows that aloha AND hawaii are more "semantically connected" then the other examples. He then shows how you can use the C-index computation to determine which engines would it be easier to target a specific keyword phrase, the higher the c-index, the more competitive that keyword phrase is in the engine, relative to other engines. Orion then explains that c-index can be used to monitor keyword trends over time, showed some very interesting slides to prove it. Orion's benchmark for a "competitive query" is one that has a c-index of above 25 points, he lists a number of those submitted to him via SEW Forums for a stufy he did several months ago. He then computed the c-index of some spam related keywords that were way above the 100 mark on the scale, neat stuff. Orion then explains that most engines use AND (FINDALL) mode as opposed to EXACT. When you look and compare both, you should find the results for EXACT mode within the FINDALL mode. The reason has something to do with order and proximity, where exact mode it does matter and findall it does not. Using this information, Orion defined a new ratio named "EF Ratio" which is equal to (n12 Exact Results/n12 FindALL Results) x 100. What the EF Ratio shows us is the "natural sequences" of words used. Meaning, how are words used in language, documents (real life). EF Ratios can be used to determine competitiveness of a keyword. The lower the number the less competitive it is. In fact, he showed that competitiveness for the same keyword phrases differ from search engine to search engine. The last slide we will save for those who were at the session.

Q & A:

LSI - Mike said that engines will use it, but he implied they are not at this time.

We spend a lot of time and money to get people to visit our websites? But what do we really know what users behave like. The room is really packed again. This looks like its going to be a great session to report on.
First up was Dr. Bonny Brown from Keynote Systems. She is going to be talking about customer experience management and how people use the web to conduct searches. About the concept of customer experience management. She asks if anyone out there has tried to interpret log files? This can be very difficult to intererpt their behavior. What are they doing? Did they not like the offer? Was it the wrong ad or audience?

To get the complete customer experience, you need a more complete picture. You need to start what the cusomter has already been exposed to. They have certain expectations. The gap is what creates the delight or disappointment. You need to understand their attitudes and reasons. Lastly the last compenent is to relate this information to making decisions. She has competitive benchmark studies that show the strengths and weakness. The mission is to help the companies understand what their strengths and weaknesses is. Understanding the why behind the strategy will help you apply it their websites. An example is vacation packages, they monitor their behavior as they navigate the website. Back to competitive benchmark studies. She took the leading search engines, and their were 2000 panelists, per site. That’s a lot of response! They have run this study twice, and see how these search engines are comparing against each other.

To get a view of how they see the behavior, is they invite panelists, and they login and search and surf the web using a browser compenent. They capture feedback and keywords they use. The customers viewpoint is that Google is #1. This is what customers are telling them. Then Yahoo #2 and MSN #3. They are seeing Yahoo and MSN closing in on Google. Ask Jeeves is also a clear success and has achieve clear gains in the space.
Ad clicking behavior is directly related to customer satisfaction. Interesting indication from the study. They do not take volume driven statistics. Ask Jeeves has resulted in the best click through rates. I am wondering if this has to do with the 10 sponsored ads they had in the natural results, which has since been tweaked.

Approx 1 in 5 users express frustration with search results relevance; product search yields fewer problems than other types of search. Some other findings from the study, say that 95% of participants use Google “sometimes or “often”. 64% use Google as their “primary search engines:. Other findings are that 3 in 4 have a primary search engines. 1 in 2 use another search engine if they cannot find what they are looking for. 1 in 3 use a search engine toolbar on their browser.

On the behavior side 3 in 4 participants began their search through a search engine or Web portal (source: clickstreams). 48% of participants used Google at some time during the task; 29% used Yahoo and 8% used MSN. 22% of people searched for a specific website they had in mind. They were also able to see how much time was spent on each website. You have the search engines, but you also have eBay and Amazon. She shows that eBay kinda sucks people in. Customer experience management helps you understand the “WHY” behind customer behavior. Sites that drive clicking behavior, may not drive a satisfying customer experience. People want relevant search results – marketers can be a partner in that goal.

Next up was Gord Hutchkiss presents some new findings about possible influences that can help us understand searcher behavior. Was hoping to see how different types of people search. They went through thousands results and manually coded them. The rank and page position were the only things that matter to users. They considered t values greater than 2. Of all the click throughs they monitored 51% clicked on the number 1 listing. There was a discrepancy between how searchers behaved as opposed to what they told us.

They created a theory, and found that a users confidence level gradually went down if they found it harder to find something they were looking for. What’s the best way to find out what they are seeing. Eye-tracking. It gives them a map of how someones eyes move across the page. Very interesting, he put up a page that showed the eyes move. These are high level finding. What was discovered was the behavior indicated that the activity on the search page show up at the top of the page. What they found that the majority of the activity happens right above the fold. “Searches golden triangle” meaning the location on the page where most people see and look at. It’s the Park Place and Broadwalk of the search page. The study also showed that the #1 paid listing also gets a high percentage of activity. If you move down beyond the page you are determined and educated about finding something down there. Back to the triangle. If the listing catches the attention to the user they will read the first listing. If not, then they scan the right side where the paid listings are. This pattern will be repeated down the page. About 50% of people are going to scan down below the fold, and the process starts over. So this scan pattern, what does it mean when building listings in the vertical space.

You need to put something that will pop and catch attention. If you do so you will get a better chance of a click through. For visibility, there is a drop down after listing number 3. After listing 8 there is a huge drop drown again in scanning behavior. Seems to level off 20-30% down at the bottom (9th and 10th) positions. 28% of people click through on the #1 paid listing. It drops off below the 3rd paid listings. If they can see the ads, why does it drop. When people where asked about how they click, they said they would click on the first thing that interested them. Asks how many people have seen rankings slip down on less popular terms. He mentions something about using click popularity as a part of how they rank. They capture how they interacted on the site. If you are above the fold you have a good chance of getting someone to click. If they are in the search triangle then they will click on the first listing. Incredible presentation!

Next up was Cam Balzer, and is discussing search before purchase. He goes into some company information. There goal was to look at the behavior up to the purchase and several areas they wanted to dig deeper. What kinds of search activity precedes the converting search/click? Search ROI usually considers only the last click before purchase. They identified 30 ecommerce website that had enough traffic to give them some reliable data. The looked at buyers and how they behaved. The findings of their study, one of the most powerful was roughly half of all buyers made a relevant search before their online purchase. They also found that marketers have a way to reach users at each step of the buying process. Travel for example had a high volume of people that did 10+ searched to find what they were looking for. Another finding was that majority of buyers searches and clicks are on generic terms. Brands do well in search, and they wanted to see how brands performed with other searches. There is a lot of people searching on generic terms to find what they were looking for. The other finding was that 30% of searchers only used a brand search. It was curious that so few users only used a branded search. Another findings, if someone was searching for a winter jacket or something in the spring, they may spend a long time in order to complete the process, sometimes up to 4-6 months. Another finding related to how the investment put into paying for generic terms also resulted in positive return for the brand name.

Room is really packed, and I am passing out yo-yo’s. Barry is also attending this session and will be covering it. So we will probably have to compare notes at the end.

First up was Jason Lamberti from comScore Networks. He runs the search marketing group of his firm. He is going to cover various segments of the market. How does comScore get their data?. They use passive tracking of actual consumer search activity. There are 1.5 million member panel of online consumers who have agreed to be continuously and passively observed. There are all kinds of data sources, and they cover the marketing thought a variety of angles. He goes into the search consumer development by country. The intensity and penetration in the U.S. lags major behind EU markets. Google dominates worldwide. You need to go through the Google network in order to market to those in countries besides the US. Search is a key marketing method to reach the working user. Overall search growth of 22.4% is driven by 25.3% increase in work audience. U.S. search and share trends have fluctuated in several areas. Recently there have been a flat amount of growth in certain areas. Reason, each search engine comes out with the same thing (desktop search, etc..) Branding is going to take a critical role in which search engine they use. He gives the example of William Jung and Ask Jeeves. Obviously we all saw this commercial. He goes on to say that search engines each have its strengths.

Local search trends, there is a conservative view of local web search suggests huge opportunities exist for marketers. Why are people searching locally? For the vast majority of the time, its searching in the area that they live, nearly 60% of local searches are for this reason. He mentions that searchers want a convenient and integrated experience. There are some reason why you might want to use a toolbar. Trial is strong – 58% of searchers have installed a toolbar. 12% of these users have uninstalled the toolbar. There is also a group that don’t even use the toolbar at all. Heavy users dominate search. There are 20% of users contribute 68% of the volume of search daily. There is the 20/30/50 rule. That 20% of searchers (meaning actual people) take up 68 percent of searches. 30% take up 24% of searches, and 50% of searchers only take up 8% of the searches. Meaning more sophisticated users do more searching and more frequently.

He next asks what users think is more important for a search engine to consider? Privacy ranks the highest as most important. When asked what would make them switch search engines? A large amount said they would switch if the relevancy is better. Jason mentions that with no barrier to switching, all search engines are vulnerable to something better or more convenient. Switching is very common and easy to do on the internet. Some 52% of Google searches are conducted on other places besides the Google site. It still contains a wide open market, and something can change.

He repeats that 85% of online purchase conversion occurs in subsequent user sessions. Consumers have a long buyer cycle, with a great deal of conversion occurring after week 4. He says this presentation is not online as the data is very valuable. I was impressed with the data and the level of coverage. If you are reading this, consider yourself privileged.

Next up was Bill Tracer from Hitwise. He wants to talk and give an overview of search landscape. On to the data, he put up a graph of the market share of the search engines against all internet sites. He talks about stats that pertain to search volume. Search volume indicates that Google is currently driving 55.5% of all U.S. search traffic. Bill next presents some click stream data, particularly the upstream traffic. The data shows that there is a clear difference between Google and Yahoo/MSN Search. Both Yahoo and MSN search benefit from portal parents. Could origination of visits explain difference in search yield (i.e. convenience search v. search mission). He mentions something called the “Mom Factor” He gives the example of his mom, and how she will go to MSN and type in www.google.com in the search bar. He then tells her she can do this in the url bar, and save the time. He asks why she did this, and she answers the internet is a big scary place. I guess you have to be here to get the joke.

He mentions psychographics of local search. Psychographic analysis of visitors to local search reveals differences between Google Local and Yahoo local search. Google Local: a slight skew to smaller second-tier cities. He then goes on to mention desktop search versus main search engine reveals a skew towards mature users. Yahoo Desktop search demonstrates the strongest skew in segments M1 and M3 (being the level of age and maturity). He mentions that older (50+) use desktop search more often. Reason: Older people loose things more often.

Next up was the Kenneth Cassar from Nielsen/ Net Ratings. He starts off by saying that innovation is not an option. He wants to imply that innovation is imperative. Covered in this session there will be the state of the competitive market in the search space. Where is tomorrow’s growth in search supply going to come from. Google continues to enjoy dominance in the search marketing share. Google takes up 47% of the market. So what about searchers? Google covers 28.7 million searches, Yahoo 13.7 million, and MSN 12.2 million searches. He then overlaps the charts and shows that there are a lot of overlapping between the various engines. One person might both use Google and Yahoo. I completely agree with this assessment. Why not use more than one? Example is that Google and Yahoo share 18.3 million searches. In the total universe of daily searches is 102 million between the three main players.
So where will tomorrows growth in search supply come from? I am enjoying Kenneth’s presentation, very well organized and clear. He describes how they looked back two years. Even though the markets has grown 100% each year, the amount of pages being added hasn’t been as much. These are interesting statistics, and very good for search marketers to hear. The online audience has been entirely responsible for the new growth year to year. A look back in history. There has also been historical threats of supply storage, innovation will create new supply. There are various segments for incremental growth. They are: advertising, new kinds of advertisers, and local search. New advertisers that are brought in the search fold. There are a several significant groups of advertisers have not yet embraced search advertising.
Local he says is one of the single best opportunities. Innovation is clear, Yahoo has developed many new innovations on this front. A9.com is an example. He says that local search opportunity is a nascent one.

They moved us all from a smaller room to a larger room and it is still a bit small. So they are moving walls around to make it even larger, movable walls - nice. Danny reused his normal joke about committing to the session and "we will commit to you" and people laughed (I guess we have newbies here).

Leading off is Dr. Bonny Brown from Keynote Systems (from Vividence), who will talk from the customers perspective. Keynote's mission is to improve e-businesses worldwide by focusing on the technical aspects and customer experience management space. She explains that there is often very little insight when tracking users as to where people get lost during the clickstream. To get the "complete customer behavior" you need more detail. Customers have expectations, and you need to understand and measure them. Then you need to understand the "whys" actions of behavior. Keynote measures industry metrics to get a better picture for benchmarks. They invite panelists, they logon, they ask you to search the Web with a toolbar (which does all that fun tracking). The results... Google is number one from a customer's experience ranking. Yahoo is number two and MSN is number three BUT Yahoo and MSN are both closing in on Google. She says, Ask Jeeves it a clear success by showing substantial gains. Ad clicking behavior seems to be INVERSELY related to the above numbers (search experience). Just keep in mind that Ask Jeeves gives you very little choice but to click on the sponsored ads (rumors are that they are going to limit that). You find a lot more frustration with searchers on complex searches. 95% use Google sometimes or often, 64% use it as their primary engine. 1 in 3 are using a toolbar. 17% use different search engines for different types of searches. 92% said they used the engines to find products. Relevance is #1 factor in search loyalty, including sponsored results. On the behavior side, they are able to see what people actually do. 3 of 4 of the people who they sat in front of a computer and asked them to find information used a search engine (wow, 25% didn't use an engine). 48% Google, 29% Yahoo!. They also measured the time they spent on the sites. Google, Yahoo, MSN, Amazon, eBay in that order are the time spent on those sites but eBay has a lot more pageviews then Google.

Gord Hotchkis from Enquiro was going to share some eye tracking studies with the help of Did-It (Kevin Lee). He explained the possible influencers include type of user, presence of brand, trusted URLs, demographics and trusted sources of information. And they expected to see a strong correlation between these sources. But what they saw was that the most important influencer was ranking. Of all the click throughs on organic listings, 24% on number 1, 19.5% number two and 12.8% on the number three listings. They came up with a theory named, the search confidence theory, which means, you trust the engine you use to give you relevant results. As you hit that back button, throughout a session, you lose your confidence level. So as you hit that back button and go to new results, you expect them to care more about the influencers listed above as opposed to just ranking. So to prove this theory, they did some eye tracking studies. They ran 50 people through in their labs, they aggregated this information at this time. He showed a slide of red, orange, blue, black, etc. The bulk of the red orange is at the top left, right where the top three results are. The majority of the eye activity is at the top, they called it "Search's Golden Triangle" which appears to include the sponsored listings and the "hidden tabs". The number one listing gets a high level of activity. The eye goes first to the top left point, then it moves to the right, to read that one listing. Then they scroll in a vertical line down on the left side, if any of those results catch someone's attention they move right again. If they don't find anything, 60% will scroll down and continue the pattern - the other 40% will look at the sponsored listings on the right. This study was only performed on Google. So you need to put something at the beginning at the title, that will catch their eyes (@#%$$). In the "golden triangle" visibility is huge up to ranking 3, then you drop off to about 60% in 4th position, and then 6, 7 and 8 go to 50% and then at 8, 9 10 drop big time 20%. Now clickthroughs, number one position 28% CTR, and then on #2 12%, and #3 11% or so, and then it drops to 6% for the remainder. So why aren't everyone clicking on the top three (numbers two and three have a much worse CTR then #1)? They click away. Sponsored results differ that the first two at the top are high visibility, and on the right side, it is very low visibility. He then summarized and told us what he wants to do more with these studies.

Last up, Cam Balzer from Perfomics (DoubleClick just bought them in July 04). They looked at search activity that lead up to the purchase or transaction. Most search tracking just looks at the last click before the purchase and most people look at the same session value (latency isnt being used). The methodology was to use ComScore's panel, identified 30 e-commerce sites in four verticals, and they identified all the buyers on the panel for a 30 day period and finally they weeded out all the random searches done. They basically weeded out a ton of irrelevant searches. They found that there are a lot of people using search before they buy. In all four verticals about 50% use search before they buy. Marketers have several opportunities to reach buyers, around 6 - 12 searches per user (interesting). They also looked at how brand keywords perform versus generic terms. Majority of search activity is around generic keywords all through the buying process (triangle searching?) Most buyers never search on a merchant brands, less then 30% on the computer vertical searched on a brand. Branded search activity peaks immediately prior to purchase. But on the computer vertical the brand search peak is much lower right before the purchase as compared to the other verticals. Almost 55% in travel space, did their last search almost two weeks before they purchased something (two weeks!). People who buy off a brand term normally start off with a generic search, but those who buy off a generic search don't normally search brand in the process. He says search is getting more competitive, to capitalize on this opportunity, you need to fully understand the value of search (not just a two week window).

This room is packed, people standing out the doors, siting on the floor and hanging from the ceilling. Three firms are up on the panel who collect data all across the Web on search related topics.

First up was James Lamberti from Comscore Networks and he runs the search group there. They get their data from 1.5 million consumers, who have agreed to be continously and passively observed. They collect all search activity, actual purchase, and other transactions. They capture everything. The data sources are many and he didn't go through them all. Search development by country, the US is lagging behind in terms of searches per searcher available. Meaning, there are more internet users in the US, but as a percentage of those users, less are using search then other countries. Google dominates around the world, but in the US, Google does not totally dominate. 22.4% search growth from Jan 04 - 05, but Work consumer group has frown 25%. Google and Yahoo have gain share at the expense of AOL, within the US. But since May 04 or so, everything has pretty much remain constant in the search area. He then broke down the share of market by segment, School: Google 42%, Yahoo 32%. Work; Google 37%, Yahoo 30% and MSN 19%. Local search trends show a slow growth from 8.3% to 10%, meaning searchers using Google and adding in state or local parameters in the query. He points out that most of it is happening at many search engines and not at yellow pages, etc. Growth in toolbar searches have grown 136% in the US since Jan 04. He said 17 - 18% of searches are coming from toolbars, WOW. 58% of searchers have not installed the toolbars and 12% have uninstalled the toolbar. In search there is a 20 - 68 rule, 20% of the users contribute 68% of volume. Heavy searchers are less likely to click on the sponsored links (27%). He noted that there is "impression value" in having your result come up in the sponsored links. 22% of the people who do not search represent 4% of the buyers online (hmmm). Satisfaction is high from most search engines. Most important search attributes for search engines at this time is privacy (Google toolbar). When you ask them, what will make you switch, the top answer is "relevancy". 49% of the searches powered by Google is happening outside of Google, which is interesting. He notes that this is still a wide open market. A vast majority of converts, are done offline, meaning lots of people doing research online will buy offline. So its very important to be able to somehow track offline conversions (a session on that later).

Bill Tancer from Hitwise, and he introduced himself as "I Love Data". He called himself a huge geek, he said finally numbers and data have become cool. They monitor the largest worldwide sample of internet users. They categorize them into 160 cats. Market share of visits to all internet sites on the Web, Google 3.1% Yahoo 1.7% and MSN 1.5% - Google has grown the most. Search volume indicates that Google is currently driving 55.5% of all US Internet searches, Yahoo 30.8% and MSN 6.64%. He gathered some clickstream data, its interesting to see that most of the search traffic is coming from the parent (yahoo.com, msn.com) but Google's upstream is coming from a wide share %, which differs from the rest. The top search from Yahoo and MSN are more navigational (www.site.com) and Google is not like that (more internet savvy). The demographics between the engines are very close, Google has a slightly higher male pop and higher income pop. He is popping up some really cool and informative slides, wish I can type faster. Google skews slightly to smaller cities compared to the other engines. Google Desktop vs. Yahoo Desktop, there is a strong skew towards the mature users, the 55+ group are more likely to use Desktop search. Verticalization of Search, will the shopping engines, etc. cannibalize the main search business. Health and medical search verticles have gone up 10% and 45% come from the main engines. 10.23% of the searches from Google go to shopping and classified sites of that eBay is the top, then amazon, etc. But when looking at other engines, it was extremely similar in the percentages. The theory is that over time, people will start have brand identity with these vertical engines and take away some traffic from main engines.

Ken Cassar from Nielsen//NetRatings was the last one up. Google's market share is at 47%, Yahoo at 21%, MSN 13% and AOL at 5%. They define a search as someone entering in a query and pressing the enter key. Google also has the largest audience of exclusive searchers; 29.7m. Yahoo has 13.7m and MSN has 12.2m and then he overlapped them (which is nice, there is an ton of cross usage of the engines). The search boom has been more driven by advertisers than consumers. Demand is not the main reason for growth, it is the supply pushing it up. This all highlights the need for innovation, in his opinion. There has been many people who talk about running out of supply. He feels its more about building more cost to switching, that should be interesting to see. MSN is advertising, Yahoo! was doing ads on local offerings and Ask is doing some ads. He said several significant groups of advertisers have not yet embraced search advertising, including; high consideration brand advertisers and low consideration direct marketers (oh so true). Is there a brand impact on people seeing paid listings? They did some tests and they saw a statistically signification different with paid listings. He won't go as far to say that brand building is better done with paid listings then through TV spots (would you?). He believes local is the single biggest opportunity there, and he points out Yahoo!. He does a query and shows some nice search results from Yahoo. A9 is an other engine focused on this he says, "Click to Call Business" (pay per call?). Local searches 3% versus all other searches 97%. They define local by using a local property, unlike Bill's or James figures which use the query term to determine local searches.

Niki Scevak from Jupiter Research opened up the Search Forecast and Outlook: Profiting from Growth Through 2009. I have attended this session about 2 times before, as I believe it provides some of the best information on the first day of the conference. He mentions that by 2009 paid search growth remains strong but beginning to slow eventually. Search will reach 5.5 billion dollar industry by 2009.

He next goes into the elements of paid search, with the first being pricing. What is seen is the increasing cost per clicking will be driving spend up and ROI down. The willingness and confidence of marketers to spend is the return on investment that paid search can return. He mentions that paid search cost will eventually stabilize. As retailers, travel merchants, and online merchants improve conversions rates some of that price is passed down as a higher CPC price. The primary aim is transactional, and if you peel away the effect, the primary goal is to produce successful sales.

The other driver of search dollars is the growing segment of households adopting broadband access. Search is the second most ambiguous activity next to email. As broadband increases from 27 million households to 46 million households will mean that users are spending more time online. This is a conservative outlook, and mentions that the forecasts for last year were short of what was real. In terms of the impact of broadband, the convenience is one of the things that will provide benefit to the people using it. Having access more frequently will allow more people to easily go online.

Relevance is improving and because so will help users make better decisions. Niki next goes into Local Search. He sees that in 2004 local search was a 408 million market, and by 2009 it should cap or exceed 879 million. This market is a long standing challenge that many companies have invested heavily to realize the full impact of this market. He says this market is in its very early stages, and there are barriers to adoption by users. One of the things they are very bullish on, is vertical search. There are many broad based online properties such as a lot of major online magazines and news sites. Search engines will launch around these categories in order to bring a broader selections of options for users. He mentions an executive survey they do yearly to investigate the amount of search spending tied to online transactions. The primary goal of the search spending is online transactions, and the dollars spent is tied directly back to instant online sales. Where there are transactions are going on, paid search marketing will be eventually tied to its success. They conducted a study that indicated that select categories (travel, media, entertainment, financial) are taking 80% of search spending seen today. Examples of these companies include Bizrate, Monster.com, and many other companies under these categories.

So what will the future hold for these categories? He says that new search engines will launch and that existing players will restructure their pricing model based on CPC and lead based acquisitions. There are a number of players that are gaining traction in the vertical marketing. 1/3 to 1/5 of users of vertical search engines come from broad based search engines such as Google and Yahoo. The role of vertical search is to further allow better searching for the user and make the transactions easier in the change. Often the users find better information and the retailers get better quality leads. We may see more companies spending more money on vertical search engines. Vertical search will not replace Google or Yahoo or one of the broad based search engines. Its an added layer to the search experience. If you go to Yahoo and type in “digital camera” you might end up on Amazon or Shopping.com and then back to Amazon to make your purchase. The process is not complete yet.

Niki next goes in a forecast of search that will profoundly impact display advertising. The advertising placement is expanding and there is a critical mass of advertising available. They have a large amount of reach, but very few frequency in terms of behavioral marketing. Its hard to measure the exact behavior. Will they use Google or Yahoo. The decision to search for a product or service can last for several months instead of just a short amount of time. He then ends and opens up for questions.

Q: How much are search engines are getting into display advertising? Where is mobile search headed?
A: Once the inventory of search sites becomes mature, where people can plug into a network. Example is Friendster, where they have a huge inventory but little behavior tracking. They are willing to share their revenue to reach a better display. If someone searches for a car, car ads may appear on the Friendster network next time they log in. Mobile search is increasing, Yellowpages is one example as it appears he is implying this is the one way people might use it. He also mentions that mobile search is not that convenient yet.

Q: What the limiting factors facing search?
A: The maturity of the online population is an issue. The improvement per months will be based on new broadband users, and one that is tenured. He says that in coming years, cost will be stabilizing, not that they will not increase, but it will be slower. As long as conversions improve, so will cost of advertising, as their many industry where there are still undiscovered keywords and inventory. This has also contributed to the large amount of growth, and it may become more flat as the inventory plays out over the next few years.

He mentions an interesting fact from the yearly survey they do, is that surveyed individuals have said that print based yellowpages (big yellow book) is more effective and easier to use than online listings or yellowpages.

Q: Where are the assumptions behind the growth through 2009 coming from?
A: Around 23% of queries have a commercial intent. This has been trending slightly upward. There are significant efforts by search engines to identify their organic results. The relevance between paid and organic should not change dramatically. He says that organic results offer more opportunity to describe more information about the site. This information helps drive more people to use organic listing, whereas paid search is limited to a certain amount of words in the copy.

Q: If Google examines the people that use paid search and only lets in the most relevant, will this improve conversions?
A: Niki says that you can not restrict the view just to a US market. Within the last 18 months, companies have tripled their budgets for paid search as they become more sophisticated.

Q: Someone in the audience mentions that paid inclusion is included in the forecast and asks why this is?
A: Yahoo is one of the players still using paid inclusion. This is evolving into some kind of transaction form. I don’t know what he means here. It’s a niche opportunity. He says that paid inclusion is certain misunderstood. Where paid ranking vs. the guarantee that your site will be spidered daily. People may be willing to paid a certain amount of money for this certainty that their new products will be spidered.

Q: Question on budgets
A: People should not look at how much should we spend? The expectation level is unreal. When you have the contextual pricing tied to search advertising. This is confusing for advertisers. I do agree!

Q: Contextual advertising trends, where are they headed?
A: Direct sales forces are becoming better. Contextual advertising networks are constrained by the success of the direct sales force. This is on one side. There are remnant inventories, based on a yield management surveys, that advertisers are looking at. It’s hard to say that basing the ads on what is on the page of the website doesn’t tell much about conversions or trends. I am sure he is directly talking about the contextual properties that are under Google’s and other search engines control (eg. Adsense, Content Match, etc..).

Yahoo turns 10 on March 2nd! Celebrating 10 years of an amazing success that has taken the internet community by storm. I don't know what 10 years in search engine years translate to people years, but I expect their might be a few grey hairs here and there. :) Just thought I would quickly mention this as I just received an email from Yahoo (as did probably many people) that they are planning a surprise on their homepage on March 2nd. They quote saying "On our birthday, we're offering you more than great content. Visit Yahoo's homepage for a special treat." Yahoo is also throwing a shindig of sorts here at the SES conference here in New York. Yahoo does know how to throw a party, so I expect it to be a lot of fun.

Check out the birthday card Yahoo has created for their anniversary.

Ben and I thought it would be fun to hand out these funky yo-yos that light up when you use them. The slogan of the Search Engine Roundtable, "The Pulse of the Search Engine Marketing Community", is imprinted on the yo-yos. Since the SEO world is filled with such ups and downs, we thought it would be clever to give out yo-yos. Then we added the glowing aspect to the yo-yo to make it "pulse" a bit.

I admit, its not as cool as having William Hung at a party, but we are not on the same level as Jeeves.

So if you see Ben or myself at the conference, please ask for one or two. After the conference, I will probably mail out a bunch to some dedicated readers. Any left over, might get some at the Toronto SES show.

Well, the SES NYC show is coming to my home town Monday, and Ben and I will be providing live coverage. Ben and I will try to attend different sessions, this way we can provide the widest coverage possible. Also, there will be a few guest authors attending the conference, who I will bug to pitch in as they can. Our goal, to provide the most detailed and unbiased coverage of this SES show.

Here are the sessions currently on my list to see:
Monday: Search Forecast and Outlook: Profiting from Growth through 2009; The Search Landscape; Searcher Behavior and Search Algorithm Research & Developments.

Tuesday: Jerry Yang Keynote; What Is Spam?; Indexing Summit; News & Webfeed Search; and maybe Branding Tactics For Search (I probably will have to leave early this day).

Wednesday: Search Convergence; Local Search Marketing Tactics; Brand Summit: Life After Google-Geico; 3:45p - 5:15p Undecided; and Evening Forum With Danny Sullivan (anything you want me to ask?).

Thursday: What Is Content?; Integrating Search Into Other Marketing; Final session undecided.

Any requests, feel free to make them. We will see what we can do. Thanks for reading everyday!

Danny Sullivan's round up article on Googles's Autolink experiment is fantastic and a must read for anyone who owns a web site and cares about the experience of their visitors.

In Google Toolbar's AutoLink & The Need For Opt-Out he writes,

"Threadwatch describes a JavaScript blocking solution cooked up by Search Guild. Download the solution (instructions are provided), insert it into your web pages. The same Threadwatch thread is also tracking any new solutions that come up -- some new server-side ones have just been posted.

Meanwhile, an anti-anti-AutoLink option appears to also be out there for users who want to override publishers trying to prevent AutoLink. I say appears because it seems like a clunky workaround that I can't really understand -- and looking at the comments posted, some others don't get it as well.

I mention it mainly because it highlights how quickly things have become absurd. You have third-parties working to prevent AutoLink and potentially others working to prevent preventing AutoLink. It's a mess."

This "mess" is born out of the fact that Google is experimenting on our web sites, and in some cases, possibly doing harm. For every parent who has a kid who has a web site on the Internet, the fact that a Google Autolink can now present a map to your house, if that web site has your address in the content, is the stuff nightmares are made of.

From the usability standpoint, removing the right to control the experience of your web site visitor is cause for concern. If someone didn't hyperlink a book, there may be a reason for that. Google doesn't ask you. The Autolink will take your visitor to Amazon anyway. Away from your web site.

I have much to say about this in Danny Sullivan's Review of Google Toolbar's AutoLink

Won't bore you with it all here :)

Monday night, Ask Jeeves is hosting an invite only party in NYC. At this party, I am told, William Hung will be making a special appearance. There are now other rumors spreading that William Hung and Jeeves have a thing going on.

I just had to post about this funny looking forum avatar over at Cre8asite Forums, member's screen name is ninjacat.

I noticed this while reading this nice thread named Froogle experience?.

Just got notice that SEMPO elected a new board today.

The thirteen Board of Directors that will serve a one-year term, starting in mid-March, are (in alphabetical order):

Ron Belanger, Carat Interactive
Chris Churchill, Fathom Online
Barbara Coll, Webmama.com
Koichiro Fukasawa, Wasabi Communications
Gordon Hotchkiss, Enquiro
Kevin Lee, Did-it.com
Mauro Lupi, Ad Maiora SpA
Jeffrey Pruitt, iCrossing
John Sanchez, Zunch Communications
Jessie Stricchiola, Alchemist Media
Julienne Thompson, Advertising.com
Dana Todd, SiteLab
David Williams, 360i

See you at SES NYC new SEMPO board.

I am sure you have seen this, where you come from a Google search and land on a page and you notice the the keywords you searched on are highlighted within the context of that page you land on. If you haven't try this search phrase, click on this search phrase then click on the first result, it should take you to DigitalPoint's forum and highlight the words.

Now what if you used a variation of this and instead of just highlighting the keywords, you insert them into the headline of the page. Why would I want to do that? Well, there have been tests done that show when you have the keywords in the landing page, it will ultimately lead to a high conversion rate. If think deeper about this and maybe the dynamic insertion of keywords into the page will lower this conversion rate in the long term. Anyway, it can be done.

There are two issues that come to mind:

(1) What if the search query is not in proper grammatical sense? Your page will look weird. Misspellings, offensive words and so on.

(2) The second issue is that the page is, in a sense, cloaked. Because the search engines can not retrieve the same page, exactly with the same context, as the user.

These issues are now being discussed in a Search Engine Watch thread.

Moderator, AussieWebmaster, reports that Google is to change the Google AdWords user interface tomorrow. He says that "Google has reported that its Adwords' interface will be unavailable on Friday February 25, 2005 from 9 pm - 12 am pst." Typically, these things happen before SES conferences. Stay tuned for more information. Forum thread at Search Engine Watch Forums.

The resident PhD, Orion, over at the Search Engine Watch Forums, provides an insight into search engine optimization that many SEOs do not think about. In his latest thread named What is Keyword Competitiveness?, Orion says that two popular keyword analysis techniques used by SEOs are "based on speculations" and is an "exercise in futility."

The two methods he dislikes include:
(1) Combining Google and Overture keyword volumes and
(2) Building a composite metric from keyword tools that have data from "dissimilar meta engines". Well most of the tools we have use "dissimilar meta engines".

I'll quote his explanation and then you can battle it out at the thread.

Combining two different or more metrics, some representing document counts and others representing query volume from dissimilar databases (Google with Overture or several meta engines), seem to be an exercise in futility: e.g., two dissimilar analytics from two different stores are combined and taken for a fair metric. Surprisingly, many SEOs/SEMs use and defend this approach, even when the arguments are based on formulas made out of thin air. Purely and simply: based on speculations.

Announced just yesterday, Brett Tabke, owner of WebmasterWorld, a forum that gets lots of coverage over here, will be holding its next conference in New Orleans on that dates of June 21-24. I made a new category for this conference, I think I will be able to attend like the WebmasterWorld Las Vegas 2004 Conference and provide detailed coverage.

Nick Wilson said there are rumors that the Vegas conference lost money. Others believe that it is just rumors and nothing more. Who cares? The WebmasterWorld conference is a great conference. If they lost money last time, then they will learn from it and move on. All businesses lose money on one thing or an other. Learn and move on.

The official forum thread can be found at WebmasterWorld, hope to see you there.

Google cleaned up and simplified the AdSense payment console screen. Jenstar has a detailed blog entry on this topic. She also started a thread at WebmasterWorld, that has already grown to four pages.

Jenstar notes that "The thing that caught my eye the most is the "Payment Type: Check" when you click the payment details. Might just be a sign of future payment options to come." It makes sense for Google to try electronic payments, should save them some money.

The thread at WebmasterWorld has some funny but valid comments. One such comment was:

Made me smile at the mix of languages on mine

Payment Type: Check
Cheque Date: 25-Jan-2005

The popular DigitalPoint Ad Network has reached a new level of popularity. People are actually advertising it on Google's AdWord network for certain keyword phrases. The reason they advertise it, is because they get "referral credit" for everyone they refer. So the cost per click pays based on the rankings it produces organically for the user who is paying for the ad.

I believe the keyword this individual was bidding on was good cpm banner network, the ad does not come up anymore. For a screen capture see here and for the forum thread visit DigitalPoint Forums.

I have had the privilege to be in communication with Jim Lanzone at Ask Jeeves on a fairly regular basis. In some of our email exchanges, he discussed how Teoma is Ask Jeeves. I couldn't find the right word for it, so I used "cannibalize". Cannibalize is a harsh word, but the point is, Ask Jeeves is what people know. You and I might know Teoma, and respect it highly but its time for Ask and Teoma to "establish themselves as a single entity." Apostolos one of the founders of Teoma is the "#1 guy driving the engine", Jim told me. Jim continues by saying, without Ask Jeeves, Teoma would not be where it is today, so in a sense "Teoma is really just a theory" - Ask Jeeves is the engine.

For us, as individuals tied so closely to the technology and search, Teoma is something inspirational to us. Subject specific popularity, looking at the Web as communities, hubs, and authorities - and bring back results in milliseconds - Apostolos and Tao are geniuses. In Mike's interview, in an earlier entry I wrote today, Jim said to Mike; "It is all Ask Jeeves now. As Apostolos said earlier, it was only seven people and it's now into triple digits."

This is not a major issue for Ask to deal with outside of the SEM community. But Ask Jeeves does not want to hurt any feelings within this industry. Apostolos said "So there is indeed an appreciation of the simple fact that you were able to help us." He commends our industry for helping the search engines, and Ask Jeeves, work harder to get to where search should be. He said they are currently at stage three of ten and he is very excited to be part of Ask Jeeves and the future of search. Even though that might mean that Teoma is just a theory, a theory that will remain with Ask Jeeves, Apostolos and the SEM community for a really long time.

Again, I started a thread on this topic at Search Engine Watch Forums.

We all had our suspicions that Google no longer used PageRank. Well, most of us thought Google just doesn't use it to the level they did a year or two ago. Mike Grehan says Google does not use PageRank period. He kind of implied that Apostolos said Google does not use PageRank at all. In the interview, Mike quoted Apostolos as saying, "Have they implemented PageRank? The answer is no." He goes on to explain that "The importance [of PageRank] has diminished because PageRank is just one piece of the ranking algorithm over there. The ranking algorithm is so much more complex now. And PageRank is just used when they want to break ties."

I started a thread on this specific topic at Search Engine Watch Forums.

This new interview by Mike Grehan named Mike Grehan in conversation with... Apostolos Gerasoulis and Jim Lanzone is a must read. One little tidbit in there, is that Apostolos says Google no longer uses PageRank.

I am still reading the interview, so more comments later.

When reading other blogs or sites that have links to specific forum posts, you often see that the person links not directly to the post that they are quoting, but rather the thread or the page the post is on. For those that do this, I'll show you how I link to individual posts within the context of a whole thread.

For example; Danny Sullivan had an excellent post today, in response to a members post on the second page of the thread named The Little Engine That Could - Part II.

The code looks like:
For example; Danny Sullivan had an excellent <a href="http://forums.searchenginewatch.com/showthread.php?p=35782#post35782" target="_blank">post</a> today, in response to a members <a href="http://forums.searchenginewatch.com/showthread.php?p=35599#post35599" target="_blank">post</a> on the <a href="http://forums.searchenginewatch.com/showthread.php?t=4285&page=2&pp=20" target="-blank">second page</a> of the thread named <a href="http://forums.searchenginewatch.com/showthread.php?t=4285">The Little Engine That Could - Part II</a>.

Getting the thread URL; i.e. http://forums.searchenginewatch.com/showthread.php?t=4285 is easy, just click on the thread name from the forum. Getting the page specific URL is easy as well, just click on the page number while in the thread or forum.

But getting the link to the specific post, within the context of the thread (so the single post isn't only shown), can be done by clicking the view source button. Then do a find command in the source code on the post date, in the case above, with Danny's post I looked for "Today, 07:56 AM" within the source and then it shows "post35782". The next step is to click on the "find new post" button and then replace where it says "p=XXXXXX#postXXXXXX" with the appropriate post number.

In firefox, you can view "selection source" which makes it a bit quicker.

Google announced today at the Google Blog new shortcut to find movies within the Google index. All you need to do is prefix the movie name with "movie:" and it will bring back results and reviews. Google gives one example of a popular phrase used in the Matrix; so you can do a search on movie: red pill blue pill and it will bring back results for the matrix.

Also, if you want to check whats playing local, just search on movie: Suffern, NY, replace Suffern, NY with your city and state.

Gary has a nice write up on this at the SEW Blog and there is a thread at Search Engine Watch Forums.

According to a thread at WebmasterWorld, GoogleBot is really crawling sites to the max, over the past few days. Some are reporting, "Never saw a crawl like that..." and others "unprecedented".

I got to run to a meeting, so I can not check my logs for bot activity right now.